Machine Learning Recommender high data intensity


We are building a recommender engine to be integrated in an app that, each time an API is called, will pull thousands of records from an Azure SQL database and create recommendations. Currently with Azure, the time taken to query thousands of records is slow (several seconds). Is there any way to improve this speed? Is another database provider such as GCP faster in this case? Thanks

Allen Wu

Posted 2020-07-24T18:09:34.770

Reputation: 13

most probably you need to change the architecture of the data in database, for example exploring graph databases or nosql databases. The first are specialised in handling multiple connections between data and the latter are very fast – Nikos M. – 2020-07-25T08:51:52.520

No answers